Julia HPC / Data Science / Simulation position at Los Alamos National Laboratory

I would like to advertise an open Julia programming position at Los Alamos National Laboratory, involving high performance computing, numerical simulation, and data science. It’s advertised as a postdoctoral position, but if you are an otherwise qualified candidate who does not meet the Ph.D. requirement, please contact me directly (nurban@lanl.gov) and we can discuss alternatives. The position is on-site and full-time for 2 years. (The “Desired Skills” is something of a wishlist; only the “Minimum Job Requirements” are critical.)

In addition to the online application instructions, please send your cover letter, CV, and research interests directly to Nathan Urban (nurban@lanl.gov), Ayan Biswas (ayan@lanl.gov), and Earl Lawrence (earl@lanl.gov).

https://lanl.jobs/los-alamos-nm/in-situ-inference-for-large-scale-scientific-simulations-postdoc/ECF683CB1EE14C5D9D477808A5EA5089/job/

Job Information

In-Situ Inference for Large-Scale Scientific Simulations Postdoc in Los Alamos, New Mexico

Vacancy Name: IRC76731

Description

Job Title In-Situ Inference for Large-Scale Scientific Simulations Postdoc

Location Los Alamos, NM, US

Organization Name CCS-7/Applied Computer Science

What You Will Do

The postdoc position in In-Situ Inference (Exascale Data Science) will involve creating novel workflows for data science, particularly Bayesian inference, within large-scale scientific simulations as the simulations are running (“in-situ”). The project will involve parallelizing and coupling scalable distributed statistical algorithms written in a high-level programming language (Julia) with complex physics models written in low-level languages (typically Fortran or C++).

Unprecedented increase in compute capabilities of the modern supercomputers enables extreme-scale scientific simulations and novel discoveries. Since the supercomputers are still bandwidth limited (i.e., far less data can be moved to permanent storage from supercomputer memory than can be simulated), many of these new discoveries will need to be achieved in-situ.

For extreme-scale simulations such as climate or space weather, statistical inference provides sophisticated methods for analyzing and understanding events of interest, such as extreme weather. While it is quite popular to use the statistical inference methods in a post-processing pipeline with moderate amounts of data, currently there is little workflow or infrastructure to achieve the same in an in-situ setting. The proposed research explores the idea of scalable statistics for distributed computing. An important aspect of this work is the creation of in-situ pipeline that couples parallel streaming statistical analysis algorithms implemented in Julia with a large-scale physics simulation. The postdoc will focus on the computational and numerical aspects of in-situ analysis, in collaboration with statisticians and physical domain scientists / model developers.

This project offers the opportunity to publish in the areas of data analysis, parallel and distributed computation, and statistics alongside the core physics domains of climate and space weather due to the novelty in the scale of the data handled and the results generated by the scalable statistical methods. The postdoc will collaborate with geoscientists, statisticians and other data scientists, and model developers.

Computer scientists in the Applied Computer Science Group (CCS-7) work in partnership with world-leading scientists in a wide range of areas such as materials science, environmental modeling, space and planetary science, cosmology, and analysis of complex engineered systems. High performance computing and simulation plays a large role in these applications and their analysis. The Data Science at Scale Team is a world leader in extreme scale scientific visualization and data analysis. Our software is used around the world on some of the most complex simulations run on high-performance computing (HPC) centers.

What You Need

Minimum Job Requirements:

Experience and knowledge in:

  • Data science and/or scientific computing in a high-level language (Julia strongly preferred, or Python)
  • Data science and/or scientific computing in a low-level language (Fortran/C++)
  • Parallel and distributed computing (MPI strongly preferred)

Desired Skills:

Additional experience in one or more of the following areas:

  • Coupling high-level language runtimes and code with low-level languages
  • Multithreading for performance improvement
  • High performance computing (HPC), especially numerical simulations
  • Scalable distributed statistical algorithms for big data and/or linear algebra or other numerical methods (using MPI, CUDA, etc.)
  • Creation of in-situ workflow (e.g., coupling in-situ data analysis code with simulations)
  • Statistical modeling and inference (e.g., Bayesian parameter estimation, Gaussian processes, probabilistic machine learning, variational inference, streaming data analysis)
  • Ability to work with domain scientists from climate and space weather modeling on applied scientific data analysis problems

Notes to Applicants:

To apply, please use the “apply now” button below. Please include a curriculum vitæ with a publication list—preferably with citation counts—and a detailed cover letter (guideline: 1–2 pages) that describes your research interests and qualifications and addresses how your qualifications meet the job requirements. You should provide the names of three people who can serve as references for your abilities.

Additional Details:

Position does not require a security clearance. Selected candidates will be subject to drug testing and other pre-employment background checks.

New-Employment Drug Test: The Laboratory requires successful applicants to complete a new-employment drug test and maintains a substance abuse policy that includes random drug testing.

Candidates may be considered for a Director’s Postdoc Fellowship and outstanding candidates may be considered for the prestigious Richard P. Feynman, Darleane Christian Hoffman, J. Robert Oppenheimer, or Frederick Reines Distinguished Postdoc Fellowships.

For more information about the Postdoc Program, go to https://www.lanl.gov/careers/career-options/postdoctoral-research/index.php .

Equal Opportunity:

Los Alamos National Laboratory is an equal opportunity employer and supports a diverse and inclusive workforce. All employment practices are based on qualification and merit, without regards to race, color, national origin, ancestry, religion, age, sex, gender identity, sexual orientation or preference, marital status or spousal affiliation, physical or mental disability, medical conditions, pregnancy, status as a protected veteran, genetic information, or citizenship within the limits imposed by federal laws and regulations. The Laboratory is also committed to making our workplace accessible to individuals with disabilities and will provide reasonable accommodations, upon request, for individuals to participate in the application and hiring process. To request such an accommodation, please send an email to applyhelp@lanl.gov or call 1-505-665-4444 option 1.

Where You Will Work

Located in northern New Mexico in the United States, Los Alamos National Laboratory is a multidisciplinary research institution engaged in strategic science on behalf of national and global security. Our workforce enjoys a collegial work environment focused on creative problem solving. We are committed to work-life balance, as well as both personal and professional growth. Los Alamos, New Mexico, enjoys excellent weather, clean air, outstanding public schools and quick, easy access to many top ski resorts, scenic hiking and biking trails, and to a range of recreational pursuits. Many employees enjoy the proximity to the nearby state capital, Santa Fe, which is known for world-class restaurants, art galleries, opera, and cultural events.

Appointment Type Postdoc

Postdoc

Req ID: IRC76731

Best,
Nathan


Nathan Urban (nurban@lanl.gov)
Scientist
Computational Physics and Methods (CCS-2)
Los Alamos National Laboratory
Nathan M. Urban

13 Likes

Awe man this sounds like so much fun to me. If I didn’t just start a new job and my wife wouldn’t likely leave me for applying to another job across the country I probably would. Hope you all find a great candidate and have a blast solving whatever problems you all are working on!

I’m planning to wrap up my PhD in about a year. What’s the hiring timeframe?